CN118407884A - Online testing and diagnosing method for vibration characteristics of blades of wind generating set - Google Patents
Online testing and diagnosing method for vibration characteristics of blades of wind generating set Download PDFInfo
- Publication number
- CN118407884A CN118407884A CN202410874016.9A CN202410874016A CN118407884A CN 118407884 A CN118407884 A CN 118407884A CN 202410874016 A CN202410874016 A CN 202410874016A CN 118407884 A CN118407884 A CN 118407884A
- Authority
- CN
- China
- Prior art keywords
- vibration
- damage
- wind speed
- blade
- blades
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000007613 environmental effect Effects 0.000 claims abstract description 46
- 238000005070 sampling Methods 0.000 claims abstract description 36
- 230000008859 change Effects 0.000 claims abstract description 18
- 230000036541 health Effects 0.000 claims abstract description 15
- 238000012423 maintenance Methods 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000013136 deep learning model Methods 0.000 claims abstract description 9
- 230000003449 preventive effect Effects 0.000 claims abstract description 7
- 230000007797 corrosion Effects 0.000 claims abstract description 6
- 238000005260 corrosion Methods 0.000 claims abstract description 6
- 238000005516 engineering process Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000012937 correction Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 238000012544 monitoring process Methods 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000007405 data analysis Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000012300 Sequence Analysis Methods 0.000 claims description 3
- 238000013145 classification model Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 230000002277 temperature effect Effects 0.000 claims description 3
- 238000012731 temporal analysis Methods 0.000 claims description 3
- 238000000700 time series analysis Methods 0.000 claims description 3
- 230000001052 transient effect Effects 0.000 claims description 3
- 238000011179 visual inspection Methods 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 abstract description 14
- 238000010248 power generation Methods 0.000 abstract description 5
- 230000008901 benefit Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009661 fatigue test Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Wind Motors (AREA)
Abstract
The invention relates to the technical field of wind power generation diagnosis, and discloses an online test and diagnosis method for the vibration characteristics of a wind generating set blade, wherein the test and diagnosis steps of the blade vibration are as follows: s1: installing vibration sensors at key positions of the blades, designing a self-adaptive data acquisition strategy by utilizing multi-mode sensor data, and automatically adjusting the sampling rate according to vibration amplitude and environmental change monitored in real time; s2: extracting key characteristics reflecting the health condition of the blade from a large amount of data by using a signal processing technology, and evaluating the influence of wind speed and temperature environmental factors on vibration characteristics; s3: designing a customized deep learning model aiming at wind power blade damage, extracting characteristics of time sequence data and vibration signals, identifying damage of different types of corrosion, crack and impact, and evaluating damage degree; s4: based on real-time data flow and historical trend, the early warning threshold is automatically adjusted, and based on damage prediction and vibration mode analysis, a preventive maintenance plan is formulated.
Description
Technical Field
The invention relates to the technical field of wind power generation diagnosis, in particular to an online test and diagnosis method for the vibration characteristics of blades of a wind generating set.
Background
With the increase of the global demand for renewable energy sources, wind energy is rapidly developed as a clean and renewable energy source, the trend of the wind generating set, especially the large-scale trend is obvious, the size, the weight, the materials and the design complexity of the blade are continuously improved as key components in a wind power generation system, and the dynamic load born by the blade under the complex wind condition is obviously increased with the increase of the size of the blade, so that the higher requirements are put on the structural safety and the long-term reliability of the blade;
at present, the problem of vibration of the blade is a diagnosis direction which needs to be focused, the vibration of the blade not only affects the power generation efficiency, but also can cause fatigue damage, cracks and even breaks, and the fault of the whole wind generating set can be caused when the vibration of the blade is serious, so that the maintenance cost is increased, the safe operation and the economic benefit of a wind farm are affected, and therefore, the accurate monitoring and diagnosis of the vibration characteristics of the blade become the problem to be solved urgently in the field of wind power generation.
Through searching, the invention patent with the Chinese patent number of CN113029480B discloses a blade fatigue test method and a blade fatigue test system of a wind generating set, compared with the prior art, the invention patent with the Chinese patent number of CN113029480B can accurately calculate the specific positions and weights of excitation points and matching points in advance, so that adjustment and optimization are not needed in a test stage, test resources and time are saved, the test load is more approximate to the actual load, the test result is more accurate, and the test period is shortened.
However, the test period is shortened, meanwhile, the condition that the generator blades can suffer from erosion, crack or impact damage during long-term operation in the natural environment is also required to be noted, the phenomenon is mainly observed by people at present, the whole wind generating set is influenced when the phenomenon is found, the damage of the blades can lead to the reduction of aerodynamic performance, the wind energy conversion efficiency is influenced, the generated energy is reduced, the original vibration frequency and vibration pattern of the damaged blades can be changed, the vibration amplitude is increased, and the structural safety of the whole wind generating set is threatened, so that the online test and diagnosis method for the vibration characteristics of the wind generating set blades is provided.
Disclosure of Invention
The invention aims to solve the defect that the prior art mainly relies on artificial prediction of the blade state, and provides an on-line test and diagnosis method for the vibration characteristics of the wind generating set blade.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
A wind generating set blade vibration characteristic on-line test and diagnosis method, the blade vibration test diagnosis step is:
S1: installing vibration sensors at key positions of the blades, designing a self-adaptive data acquisition strategy by utilizing multi-mode sensor data, and automatically adjusting the sampling rate according to vibration amplitude and environmental change monitored in real time;
S2: extracting key characteristics reflecting the health condition of the blade from a large amount of data by using a signal processing technology, and evaluating the influence of wind speed and temperature environmental factors on vibration characteristics;
S3: designing a customized deep learning model aiming at wind power blade damage, extracting characteristics of time sequence data and vibration signals, identifying damage of different types of corrosion, crack and impact, and evaluating damage degree;
S4: based on real-time data flow and historical trend, the early warning threshold is automatically adjusted, and based on damage prediction and vibration mode analysis, a preventive maintenance plan is formulated.
In S1, determining key areas, blade roots, blade tips, middle parts of the blades and known weak points of the blades, which are most prone to damage, installing different types of vibration sensors and environment parameter sensors at the key positions, and designing a set of algorithm to dynamically adjust the sampling rate based on vibration amplitude threshold values and environment parameter changes.
In S1, the current vibration amplitudeGreater than a set thresholdAdjusting sampling rate according to the ratio exceeding the thresholdThe formula of the vibration amplitude adjusting range is:
Controlling the slope of the adjustment for an adjustment factor within the vibration amplitude threshold range;
environmental parameter adjustment dynamically adjusts sampling rate based on environmental parameter variation, sets current wind speed Wind speed at the previous momentMore than a change in (2)Then the sampling rate is adjusted according to the wind speed change proportion:
Is the adjustment factor of the wind speed variation;
representing a base sampling rate, i.e. a default sampling frequency when no wind speed variation is affecting;
: the maximum allowable sampling rate represents the highest sampling frequency limit that the system can handle;
when the vibration amplitude exceeds a preset threshold or the environmental conditions change drastically, increasing the sampling rate to capture more detail; conversely, the sampling rate is reduced during smooth operation to save resources.
In S2, a relation model between features is established through experiment or historical data analysis, actual measurement values of vibration parameters and environmental parameters are substituted into corresponding compensation models, an expected 'environmental impact vibration feature' under the current environmental condition is calculated, the main impact of the set wind speed on blade vibration can be approximated through a linear relation, and the compensation models can be expressed as:
is a correction value of the vibration amplitude under the influence of wind speed, Is the wind speed influence coefficient,Is the measured wind speed;
Subtracting the calculated environmental impact vibration characteristic from the original vibration characteristic to obtain a corrected vibration characteristic, and performing deep analysis on the corrected vibration characteristic to evaluate the health condition of the blade.
In S2, the vibration parameter and the environmental parameter are corrected, the amplitude of the original vibration signal at the frequency ff is the sum of the wind speed and the temperature correction value obtained by the compensation model, and the corrected vibration characteristics are expressed as:
: expressed in frequency The corrected vibration amplitude is then measured;
: the vibration amplitude at the frequency ff is obtained by original measurement;
: a correction value representing the influence of wind speed on vibration;
: then it is a correction value for the temperature effect on vibration;
the vibration influence expected from the wind speed and temperature change is subtracted from the original vibration amplitude, and a predictive model is built based on the corrected characteristics.
In S3, the vibration amplitude corrected based on the corrected vibration signal and the wind speed is arranged into a time series data format, each sample includes a period of time series data and a corresponding damage state label, no damage, no erosion, no crack, no impact, marking of damage type and degree is performed on the historical data of the blade according to physical inspection, ultrasonic detection and visual inspection means, time series analysis is performed based on a one-dimensional convolutional neural network model, and the convolutional layer is expressed as:
is the function of the activation and, Is a bias term that is used to determine,Is the weight of the convolution kernel,Is an input signal;
And extracting the periodicity, the trend characteristics and the transient variation characteristics in the time sequence by using time sequence analysis.
In S3, based on the extracted features of the deep learning model, for damage degree assessment, modifying a model output layer to be continuous value output, training by using a loss function of a regression task, predicting the severity of damage, dividing the damage degree into a plurality of grades, simultaneously predicting damage types and grades, further establishing a regression model of the damage degree for each type of damage on the basis of a classification model, deploying the trained model into a wind power blade health monitoring system, analyzing blade vibration data in real time, and automatically identifying the damage types and the damage degrees.
In S4, based on historical vibration data and known damage events, analyzing the trend of vibration characteristics over time, identifying vibration characteristic modes under different damage types, setting a dynamic threshold model, outputting according to real-time data and a prediction model, dynamically adjusting an early warning threshold, setting the threshold to be out of a standard deviation or a specific percentile of a normal vibration characteristic prediction interval, evaluating the health condition and potential risk of the blade according to the deviation degree of the damage prediction result and the vibration characteristics, setting different maintenance trigger thresholds according to the risk level, wherein slight deviation from the normal range may only require intensive monitoring, and serious deviation may require immediate inspection or maintenance.
The invention has the following beneficial effects:
According to the invention, through real-time monitoring of the vibration characteristics of the blade, early warning can be sent out at the early stage of damage formation, so that preventive maintenance measures are taken, high maintenance cost and long-time shutdown loss caused by sudden faults are avoided, and early warning and prevention are realized.
According to the invention, the deep learning model is introduced for testing and diagnosis, so that the complex blade vibration mode can be learned from mass data, tiny abnormal changes can be identified, the method can adapt to different wind field environments and fan types, higher diagnosis efficiency can be maintained even when the environmental conditions change, preventive maintenance is realized, unplanned downtime is reduced, and the wind field operation efficiency is improved.
Drawings
FIG. 1 is a diagram showing the steps of a method for online testing and diagnosing the vibration characteristics of a blade of a wind generating set according to the present invention.
S1: installing vibration sensors at key positions of the blades, designing a self-adaptive data acquisition strategy by utilizing multi-mode sensor data, and automatically adjusting the sampling rate according to vibration amplitude and environmental change monitored in real time;
S2: extracting key characteristics reflecting the health condition of the blade from a large amount of data by using a signal processing technology, and evaluating the influence of wind speed and temperature environmental factors on vibration characteristics;
S3: designing a customized deep learning model aiming at wind power blade damage, extracting characteristics of time sequence data and vibration signals, identifying damage of different types of corrosion, crack and impact, and evaluating damage degree;
S4: based on real-time data flow and historical trend, the early warning threshold is automatically adjusted, and based on damage prediction and vibration mode analysis, a preventive maintenance plan is formulated.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
As shown in FIG. 1, the invention provides an on-line test and diagnosis method for the vibration characteristics of a wind generating set blade, which comprises the following steps:
S1: installing vibration sensors at key positions of the blades, designing a self-adaptive data acquisition strategy by utilizing multi-mode sensor data, and automatically adjusting the sampling rate according to vibration amplitude and environmental change monitored in real time;
S2: extracting key characteristics reflecting the health condition of the blade from a large amount of data by using a signal processing technology, and evaluating the influence of wind speed and temperature environmental factors on vibration characteristics;
S3: designing a customized deep learning model aiming at wind power blade damage, extracting characteristics of time sequence data and vibration signals, identifying damage of different types of corrosion, crack and impact, and evaluating damage degree;
S4: based on real-time data flow and historical trend, the early warning threshold is automatically adjusted, and based on damage prediction and vibration mode analysis, a preventive maintenance plan is formulated.
In S1, determining key areas, blade roots, blade tips, middle parts of the blades and known weak points of the blades, which are most prone to damage, installing different types of vibration sensors and environment parameter sensors at the key positions, and designing a set of algorithm to dynamically adjust the sampling rate based on vibration amplitude threshold values and environment parameter changes.
In S1, the current vibration amplitudeGreater than a set thresholdAdjusting sampling rate according to the ratio exceeding the thresholdThe formula of the vibration amplitude adjusting range is:
Controlling the slope of the adjustment for an adjustment factor within the vibration amplitude threshold range;
environmental parameter adjustment dynamically adjusts sampling rate based on environmental parameter variation, sets current wind speed Wind speed at the previous momentMore than a change in (2)Then the sampling rate is adjusted according to the wind speed change proportion:
Is the adjustment factor of the wind speed variation;
representing a base sampling rate, i.e. a default sampling frequency when no wind speed variation is affecting;
: the maximum allowable sampling rate represents the highest sampling frequency limit that the system can handle;
when the vibration amplitude exceeds a preset threshold or the environmental conditions change drastically, increasing the sampling rate to capture more detail; conversely, the sampling rate is reduced during smooth operation to save resources.
In S2, a relation model between features is established through experiment or historical data analysis, actual measurement values of vibration parameters and environmental parameters are substituted into corresponding compensation models, an expected 'environmental impact vibration feature' under the current environmental condition is calculated, the main impact of the set wind speed on blade vibration can be approximated through a linear relation, and the compensation models can be expressed as:
is a correction value of the vibration amplitude under the influence of wind speed, Is the wind speed influence coefficient,Is the measured wind speed;
Subtracting the calculated environmental impact vibration characteristic from the original vibration characteristic to obtain a corrected vibration characteristic, and performing deep analysis on the corrected vibration characteristic to evaluate the health condition of the blade.
In S2, the vibration parameter and the environmental parameter are corrected, the amplitude of the original vibration signal at the frequency ff is the sum of the wind speed and the temperature correction value obtained by the compensation model, and the corrected vibration characteristics are expressed as:
: expressed in frequency The corrected vibration amplitude is then measured;
: the vibration amplitude at the frequency ff is obtained by original measurement;
: a correction value representing the influence of wind speed on vibration;
: then it is a correction value for the temperature effect on vibration;
the vibration influence expected from the wind speed and temperature change is subtracted from the original vibration amplitude, and a predictive model is built based on the corrected characteristics.
In this embodiment, in S1, the influence of the vibration amplitude and the environmental parameter is comprehensively considered, and a larger value or a weighted average value of the adjustment results of the vibration amplitude and the environmental parameter is taken as the final adjusted sampling rate:
A wind driven generator blade health monitoring system is designed, an early warning threshold value is required to be adjusted according to the vibration frequency of the blade and environmental factors so as to prevent the damage of the blade,
Representing the adjustment frequency based on the blade vibration signature analysis. For example, by analyzing blade vibration data, it is found that when the vibration frequency exceeds a certain threshold (based on historical data and damage cases), which threshold may be adjusted with subtle changes in the vibration characteristics (e.g., changes in amplitude, frequency content), it is predicted that there is a risk of damage;
Is a frequency that is adjusted based on environmental factors. It is contemplated that increased wind speed may result in increased blade vibration, but such increased vibration is a normal physical reaction and is not an indicator of damage. To exclude the effect of such normal variations, we calculate an adjusted threshold based on current wind speed and temperature conditions to reflect the "normal" vibration upper limit under certain environmental conditions;
An initial early warning threshold value set after blade vibration data analysis is based on the initial early warning threshold value;
after the current wind speed of 20 m/s and the environmental impact at the temperature of 25 ℃ are adjusted, the vibration frequency of the blade is considered not to exceed the value under the environmental condition;
according to the formula Final early warning thresholdThe larger value of the two values means that under the current environmental condition, even if the threshold value based on vibration characteristic analysis is lower, the threshold value adjusted by the environmental factors is used as the reference, so that misjudgment is avoided, and safe operation of the blade in a severe environment is ensured.
Example two
As shown in fig. 1, based on the first embodiment, in S3, the corrected vibration amplitude is arranged into a time series data format based on the corrected vibration signal and the wind speed, each sample includes a time series data and a corresponding damage status label, the method has the advantages that no damage, corrosion, crack and impact are generated, the damage type and degree of historical data of the blade are marked according to physical inspection, ultrasonic detection and visual inspection means, time series analysis is carried out based on a one-dimensional convolutional neural network model, and a convolutional layer is expressed as follows:
is the function of the activation and, Is a bias term that is used to determine,Is the weight of the convolution kernel,Is an input signal;
And extracting the periodicity, the trend characteristics and the transient variation characteristics in the time sequence by using time sequence analysis.
In S3, based on the extracted features of the deep learning model, for damage degree assessment, modifying a model output layer to be continuous value output, training by using a loss function of a regression task, predicting the severity of damage, dividing the damage degree into a plurality of grades, simultaneously predicting damage types and grades, further establishing a regression model of the damage degree for each type of damage on the basis of a classification model, deploying the trained model into a wind power blade health monitoring system, analyzing blade vibration data in real time, and automatically identifying the damage types and the damage degrees.
In S4, based on historical vibration data and known damage events, analyzing the trend of vibration characteristics over time, identifying vibration characteristic modes under different damage types, setting a dynamic threshold model, outputting according to real-time data and a prediction model, dynamically adjusting an early warning threshold, setting the threshold to be out of a standard deviation or a specific percentile of a normal vibration characteristic prediction interval, evaluating the health condition and potential risk of the blade according to the deviation degree of the damage prediction result and the vibration characteristics, setting different maintenance trigger thresholds according to the risk level, wherein slight deviation from the normal range may only require intensive monitoring, and serious deviation may require immediate inspection or maintenance.
In this embodiment, the construction process of the one-dimensional convolutional neural network (1D CNN) model is as follows:
firstly, uniformly scaling the numerical ranges of a vibration signal and an environmental parameter, normalizing the vibration amplitude to be between 0 and 1, then performing sequence slicing, and slicing a continuous vibration signal into a sequence with a fixed length as the input of a model;
Input layer: the shape is the sequence length, the number of channels depends on whether the environmental parameters are added as additional input dimensions;
convolution layer: using a one-dimensional convolution kernel;
activation function: for increasing the nonlinearity of the model;
Full tie layer: a space for mapping features of the convolutional layer output to classification or regression problems, e.g., 128 neurons;
Output layer: according to the task decision, if the task is a classified task, outputting various probabilities by using a Softmax function; if the task is a regression task, directly outputting a predicted value of the damage degree;
loss function: the classification task uses cross entropy loss, the regression task uses Mean Square Error (MSE) or Root Mean Square Error (RMSE);
an optimizer: adam, for updating network weights;
The data are divided into a training set, a verification set and a test set, then a compiling model is carried out, a loss function, an optimizer and an evaluation index are designated, model parameters are optimized through a back propagation and gradient descent algorithm, and the design of a model basic framework is completed.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. The on-line testing and diagnosing method for the vibration characteristics of the blades of the wind generating set is characterized by comprising the following steps of:
S1: installing vibration sensors at key positions of the blades, designing a self-adaptive data acquisition strategy by utilizing multi-mode sensor data, and automatically adjusting the sampling rate according to vibration amplitude and environmental change monitored in real time;
S2: extracting key characteristics reflecting the health condition of the blade from a large amount of data by using a signal processing technology, and evaluating the influence of wind speed and temperature environmental factors on vibration characteristics;
S3: designing a customized deep learning model aiming at wind power blade damage, extracting characteristics of time sequence data and vibration signals, identifying damage of different types of corrosion, crack and impact, and evaluating damage degree;
S4: based on real-time data flow and historical trend, the early warning threshold is automatically adjusted, and based on damage prediction and vibration mode analysis, a preventive maintenance plan is formulated.
2. The method for on-line testing and diagnosing vibration characteristics of blades of a wind generating set according to claim 1, wherein in S1, a critical area in the blades, which is most likely to be damaged, a blade root, a blade tip, a middle part of the blades and known weak points are determined, different types of vibration sensors and environment parameter sensors are installed at the critical positions, and a set of algorithm is designed to dynamically adjust sampling rate based on vibration amplitude threshold values and environment parameter changes.
3. The method for on-line testing and diagnosing vibration characteristics of a wind turbine generator system according to claim 1, wherein in S1, the current vibration amplitude is determinedGreater than a set thresholdAdjusting sampling rate according to the ratio exceeding the thresholdThe formula of the vibration amplitude adjusting range is:
Controlling the slope of the adjustment for an adjustment factor within the vibration amplitude threshold range;
environmental parameter adjustment dynamically adjusts sampling rate based on environmental parameter variation, sets current wind speed Wind speed at the previous momentMore than a change in (2)Then the sampling rate is adjusted according to the wind speed change proportion:
Is the adjustment factor of the wind speed variation;
representing a base sampling rate, i.e. a default sampling frequency when no wind speed variation is affecting;
: the maximum allowable sampling rate represents the highest sampling frequency limit that the system can handle;
when the vibration amplitude exceeds a preset threshold or the environmental conditions change drastically, increasing the sampling rate to capture more detail; conversely, the sampling rate is reduced during smooth operation to save resources.
4. The method for online testing and diagnosing the vibration characteristics of the blades of the wind generating set according to claim 1, wherein in S2, a relation model between characteristics is established through experiments or historical data analysis, actual measurement values of vibration parameters and environmental parameters are substituted into corresponding compensation models, the expected vibration characteristics of environmental influence under the current environmental conditions are calculated, the main influence of the set wind speed on the vibration of the blades can be approximated through a linear relation, and the compensation models can be expressed as:
is a correction value of the vibration amplitude under the influence of wind speed, Is the wind speed influence coefficient,Is the measured wind speed;
Subtracting the calculated environmental impact vibration characteristic from the original vibration characteristic to obtain a corrected vibration characteristic, and performing deep analysis on the corrected vibration characteristic to evaluate the health condition of the blade.
5. The method for on-line testing and diagnosing vibration characteristics of a wind turbine generator system according to claim 1, wherein in S2, the vibration parameters and the environmental parameters are corrected, the amplitude of the original vibration signal at the frequency ff is the sum of the wind speed and the temperature correction value obtained by the compensation model, and the corrected vibration characteristics are expressed as:
: expressed in frequency The corrected vibration amplitude is then measured;
: the vibration amplitude at the frequency ff is obtained by original measurement;
: a correction value representing the influence of wind speed on vibration;
: then it is a correction value for the temperature effect on vibration;
the vibration influence expected from the wind speed and temperature change is subtracted from the original vibration amplitude, and a predictive model is built based on the corrected characteristics.
6. The method for online testing and diagnosing the vibration characteristics of the blades of the wind generating set according to claim 1, wherein in S3, the corrected vibration signals and the corrected vibration amplitudes of the wind speed are arranged into a time series data format, each sample comprises a period of time series data and a corresponding damage state label, no damage, no erosion, no crack and no impact are caused, the damage type and the damage degree are marked on the historical data of the blades according to physical inspection, ultrasonic detection and visual inspection means, the time series analysis is carried out based on a one-dimensional convolutional neural network model, and a convolutional layer is expressed as:
is the function of the activation and, Is a bias term that is used to determine,Is the weight of the convolution kernel,Is an input signal;
And extracting the periodicity, the trend characteristics and the transient variation characteristics in the time sequence by using time sequence analysis.
7. The method for online testing and diagnosing the vibration characteristics of the blades of the wind turbine generator system according to claim 1, wherein in the step S3, based on the characteristics extracted by the deep learning model, for damage degree assessment, a model output layer is modified to be continuous value output, a loss function of a regression task is used for training, the severity of damage is predicted, the damage degree is divided into a plurality of grades, meanwhile, damage types and grades are predicted, a regression model of the damage degree is further built for each type of damage on the basis of a classification model, the trained model is deployed into a wind turbine blade health monitoring system, blade vibration data is analyzed in real time, and the damage types and the damage degrees are automatically identified.
8. The method for online testing and diagnosing the vibration characteristics of the blades of the wind generating set according to claim 1, wherein in the step S4, based on historical vibration data and known damage events, trends of vibration characteristics along with time are analyzed, vibration characteristic modes under different damage types are identified, a dynamic threshold model is set, an early warning threshold is dynamically adjusted according to real-time data and output of a prediction model, the threshold is set to be out of a standard deviation or a specific percentile of a normal vibration characteristic prediction interval, health conditions and potential risks of the blades are evaluated according to deviation degrees of the damage prediction results and the vibration characteristics, different maintenance trigger thresholds are set according to risk grades, and only the monitoring is required to be enhanced when the slight deviation from the normal range, and inspection or maintenance is required to be immediately arranged when the serious deviation occurs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410874016.9A CN118407884B (en) | 2024-07-02 | 2024-07-02 | Online testing and diagnosing method for vibration characteristics of blades of wind generating set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410874016.9A CN118407884B (en) | 2024-07-02 | 2024-07-02 | Online testing and diagnosing method for vibration characteristics of blades of wind generating set |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118407884A true CN118407884A (en) | 2024-07-30 |
| CN118407884B CN118407884B (en) | 2024-09-03 |
Family
ID=91990020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410874016.9A Active CN118407884B (en) | 2024-07-02 | 2024-07-02 | Online testing and diagnosing method for vibration characteristics of blades of wind generating set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118407884B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107829885A (en) * | 2017-10-25 | 2018-03-23 | 西安锐益达风电技术有限公司 | A kind of blade of wind-driven generator vibration monitoring and system for considering ambient parameter amendment |
| CN111649887A (en) * | 2020-06-17 | 2020-09-11 | 内蒙古工业大学 | Wind turbine blade vibration characteristic testing device |
| TWI732660B (en) * | 2020-08-20 | 2021-07-01 | 國立勤益科技大學 | Wind power generator fault diagnosis system and method |
| EP4050206A1 (en) * | 2021-02-25 | 2022-08-31 | Siemens Gamesa Renewable Energy A/S | Controlling a wind turbine with respect to dynamic stability |
| CN117869224A (en) * | 2024-02-04 | 2024-04-12 | 保定市瑞高电气有限公司 | Vibration monitoring system and vibration suppression method for wind driven generator |
-
2024
- 2024-07-02 CN CN202410874016.9A patent/CN118407884B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107829885A (en) * | 2017-10-25 | 2018-03-23 | 西安锐益达风电技术有限公司 | A kind of blade of wind-driven generator vibration monitoring and system for considering ambient parameter amendment |
| CN111649887A (en) * | 2020-06-17 | 2020-09-11 | 内蒙古工业大学 | Wind turbine blade vibration characteristic testing device |
| TWI732660B (en) * | 2020-08-20 | 2021-07-01 | 國立勤益科技大學 | Wind power generator fault diagnosis system and method |
| EP4050206A1 (en) * | 2021-02-25 | 2022-08-31 | Siemens Gamesa Renewable Energy A/S | Controlling a wind turbine with respect to dynamic stability |
| CN117869224A (en) * | 2024-02-04 | 2024-04-12 | 保定市瑞高电气有限公司 | Vibration monitoring system and vibration suppression method for wind driven generator |
Non-Patent Citations (2)
| Title |
|---|
| XIN JIANG: "Comparative Study on Vibration Characteristics of Biaxial Carbon/Glass Hybrid Wind Turbine Blades", APPLIED SCIENCE, 31 August 2024 (2024-08-31), pages 1 - 20 * |
| 姜焱: "基于有限元法的三维机织复合材料叶片振动特性分析", 可再生能源, vol. 35, no. 7, 31 July 2017 (2017-07-31), pages 1066 - 1071 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118407884B (en) | 2024-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Igba et al. | Analysing RMS and peak values of vibration signals for condition monitoring of wind turbine gearboxes | |
| Uluyol et al. | Power curve analytic for wind turbine performance monitoring and prognostics | |
| EP1982301B1 (en) | Method of condition monitoring | |
| WO2018044507A1 (en) | Machine learning-aided model-based method for estimating gas turbine component life | |
| Bangalore et al. | An approach for self evolving neural network based algorithm for fault prognosis in wind turbine | |
| JP6315836B2 (en) | Windmill monitoring device, windmill monitoring method, and windmill monitoring program | |
| CN110794683A (en) | Wind power gear box state evaluation method based on deep neural network and kurtosis characteristics | |
| CN114881087A (en) | Building robot bearing performance degradation assessment method | |
| CN116384975A (en) | Maintenance system based on power plant equipment | |
| CN114004059B (en) | Health portrait method for hydroelectric generating set | |
| CN118407884B (en) | Online testing and diagnosing method for vibration characteristics of blades of wind generating set | |
| CN111192163B (en) | Generator reliability medium-short term prediction method based on wind turbine generator operating data | |
| CN116717436A (en) | Variable pitch system fault early warning method and system | |
| Ding et al. | Integrated prognosis for wind turbine gearbox condition-based maintenance considering time-varying load and crack initiation time uncertainty | |
| KR20240003469A (en) | Predictive diagnosis method and system of nuclear power plant equipment | |
| Shen | Classification of Wind Turbine Blade Performance State Through Statistical Methods | |
| CN114439704A (en) | Wind turbine generator running state monitoring method based on SCADA (supervisory control and data acquisition) and FISSA-DBN (fuzzy inference system-database network) models | |
| Obdam et al. | Flight leader concept for wind farm load counting: First offshore implementation | |
| CN114151291B (en) | Early fault monitoring method for wind turbine generator | |
| CN118137926B (en) | Frequency modulation control method for three-phase asynchronous motor based on electromechanical signal analysis | |
| CN118014405B (en) | Power failure early warning system based on data processing | |
| Ao et al. | Data-Driven Modeling for Wind Turbine Blade Loads Based on Deep Neural Network | |
| Kobenko et al. | Monitoring of the Wind Turbine Bearing Condition Using Identification Measurement Technology | |
| CN118410279A (en) | Deep learning-based turbine unit fault processing method and system | |
| CN118327910A (en) | Wind turbine generator blade state acquisition system and method based on optical fiber sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |